Proceedings Track
Full papers published in Proceedings for Machine Learning Research (PMLR), up to 9 pages plus unlimited references and appendix; must not be published or under review elsewhere; requires in-person presentation for inclusion.
Machine Learning/CORE Unranked
LOG 2026
The Fifth Learning on Graphs Conference
The Learning on Graphs Conference (LoG) 2026 is an annual research conference focused on machine learning on graphs and geometry, emphasizing high-quality peer review. It features both in-person and virtual components, with a strong community-driven approach including local meetups and a unique reviewer reward system.
Paper fit
Contribution paths
A strong submission should clearly identify its contribution and evaluate it appropriately.
Extended Abstract Track
Non-archival submissions up to 4 pages plus unlimited references and appendix; allows previously published or concurrently submitted work; welcomes novel datasets, negative results, preliminary findings, and reproducibility studies; retains full copyright.
NeurIPS 2025 Fast Track
Special track for NeurIPS 2025 submissions with average score ≥4.0; requires submission of original paper, reviews, meta-review, author response, and ethics statement; handled via OpenReview.
Tutorial Proposals
1.5-hour or 3-hour in-person tutorials on graph and geometric machine learning; must be self-contained, balanced, and include hands-on components; proposals limited to 4 pages.
Research areas in scope
01
Research Areas
Expressive Graph Neural NetworksGNN architectures (transformers, new positional encodings, …)Equivariant architecturesStatistical theory on graphsCausal inference (structural causal models, …)Algorithmic reasoningGeometry processingRobustness and adversarial attacks on graphsTrustworthy graph ML (fairness, privacy, …)Combinatorial Optimization and Graph AlgorithmsGeometric and graph generative models (Diffusion, Flow Matching, …)Graph Foundation ModelsGraph KernelsGraph Signal Processing/Spectral MethodsGraph Generative ModelsScalable Graph Learning Models and MethodsGraphs for Recommender SystemsKnowledge GraphsKnowledge Graphs and LLMsGraph/Geometric ML for Computer VisionGraph ML for Natural Language Processing and LLMsGraph/Geometric ML for Molecules (molecules, proteins, drug discovery, …)Graph ML for SecurityGraph ML for HealthGraph/Geometric ML for Physical sciencesGraph ML Platforms and SystemsSelf-supervised learning on graphsGraph/Geometric ML Infrastructures (datasets, benchmarks, libraries, …)Networks AnalysisManifold learningNeural manifoldGeometric optimizationStructured probabilistic inferenceLLMs and GraphsLLMs for Recommender SystemsLLMs and GeometryGraphs, Agents and Multi-Agent Systems
Policies worth checking twice
- Submissions are double-blind; author and reviewer identities are concealed during review.
- Authors may submit anonymized work that is already available as a non-anonymous preprint without citing it.
- Proceedings track papers cannot be published or under review in any other archival venue.
- Extended abstract track submissions may be under review or previously published, provided they do not violate other venue policies.
- At least one author of each accepted paper must attend the conference in person; exceptions require prior approval from Program Chairs.
- Rebuttal and author-reviewer discussion periods are conducted via OpenReview with public comments allowed.
- Accepted papers are deanonymized after notification; rejected papers may opt out of deanonymization.
- Generative AI tools may be used judiciously but cannot be listed as authors; prompt injection is strictly forbidden.
Official sources
Compiled from the official call for papers. The organizers’ pages remain authoritative.
Official conference siteCall for PapersCall for TutorialsProgram ChairsDownload Camera Ready Templatedownload
Last verified September 9, 2026